Trait opencv::hub_prelude::BEBLID

source ·

pub trait BEBLID: Feature2DTrait + BEBLIDConst {
    // Required method
    fn as_raw_mut_BEBLID(&mut self) -> *mut c_void;

    // Provided method
    fn set_scale_factor(&mut self, scale_factor: f32) -> Result<()> { ... }
}

Expand description

Class implementing BEBLID (Boosted Efficient Binary Local Image Descriptor), described in Suarez2020BEBLID .

BEBLID \cite Suarez2020BEBLID is a efficient binary descriptor learned with boosting. It is able to describe keypoints from any detector just by changing the scale_factor parameter. In several benchmarks it has proved to largely improve other binary descriptors like ORB or BRISK with the same efficiency. BEBLID describes using the difference of mean gray values in different regions of the image around the KeyPoint, the descriptor is specifically optimized for image matching and patch retrieval addressing the asymmetries of these problems.

If you find this code useful, please add a reference to the following paper:

Iago Suárez, Ghesn Sfeir, José M. Buenaposada, and Luis Baumela. BEBLID: Boosted efficient binary local image descriptor. Pattern Recognition Letters, 133:366–372, 2020.

The descriptor was trained using 1 million of randomly sampled pairs of patches (20% positives and 80% negatives) from the Liberty split of the UBC datasets \cite winder2007learning as described in the paper Suarez2020BEBLID. You can check in the AKAZE example how well BEBLID works. Detecting 10000 keypoints with ORB and describing with BEBLID obtains 561 inliers (75%) whereas describing with ORB obtains only 493 inliers (63%).